ORIGINAL RESEARCH PEDIATRICS

Cavum Septum Pellucidum in the General Pediatric Population and Its Relation to Surrounding Structure Volumes, Cognitive Function, and Emotional or Behavioral Problems

X M.H.G. Dremmen, X R.H. Bouhuis, X L.M.E. Blanken, X R.L. Muetzel, X M.W. Vernooij, X H.E. Marroun, X V.W.V. Jaddoe, X F.C. Verhulst, X H. Tiemeier, and X T. White

ABSTRACT BACKGROUND AND PURPOSE: The cavum septum pellucidum, a cavity filled with CSF, is localized between the 2 of the brain. The cavum is present in all neonates, but it typically closes within 5 months after birth. In some cases, this closure does not occur and a persistent or enlarged cavum septum pellucidum has been linked, in some studies, to psychiatric disorders. However, the clinical relevance in the general population is unknown. In this study, we examined the relationship between the cavum septum pellucidum and volumes of brain structures, cognitive function, and emotional and behavioral problems in children.

MATERIALS AND METHODS: This study was embedded in the Generation R Study, a prospective cohort in Rotterdam, the Netherlands. MR imaging studies of 1070 children, 6–10 years of age, were systematically evaluated for the presence and length of a persistent cavum septum pellucidum. An enlarged cavum septum pellucidum was defined as a cavum length of Ն6 mm. Groups without, with persistent, and with enlarged cavum septi pellucidi were compared for brain structure volumes, nonverbal intelligence, and emotional and behavioral problems.

RESULTS: The prevalence of cavum septi pellucidi in our sample was 4.6%. Children with an enlarged cavum septum pellucidum had a larger , greater thalamic and total –to–total brain volume ratio, and smaller lateral ventricle volumes. We did not find a relationship between cavum septi pellucidi and cognitive function or emotional and behavioral problems.

CONCLUSIONS: The cavum septum pellucidum is a normal structural brain variation without clinical implications in this population-based sample of school-aged children.

ϭ ϭ ϭ ϭ ␳ ϭ ABBREVIATIONS: CBCL Child Behavior Checklist; CSP cavum septum pellucidum; IQ intelligence quotient; M marginal mean; s Spearman partial correlation; SE ϭ standard error

he septum pellucidum is a thin plate consisting of 2 fused the surrounding brain structures.1-3 However, in some cases, the Tlaminae or septa located between the lateral ventricles of the septa do not fuse and form the persistent cavum. A persistent CSP brain. At birth, these 2 septa are separated and form the lateral with a cavum length of Ͻ5 mm is common in a large proportion walls of a cavity filled with CSF, the cavum septum pellucidum of healthy subjects, with prevalence rates of up to 30% in the adult (CSP). The 2 septa of the cavum typically fuse into a single septum population,1,4 and is therefore considered a normal variant. In pellucidum within 5 months after birth, likely due to growth of general, the size of a normal-variant septum is estimated at ap- proximately 1–4 mm.1,5 There is no clear cutoff value to define enlargement of the CSP, but many previous studies used a com- Ն 2,3,5-16 Received August 29, 2018; accepted after revision December 1. mon cutoff of 6 mm. From the Departments of Radiology (M.H.G.D., R.H.B., M.W.V., T.W.), Epidemiology In patients with schizophrenia and an enlarged CSP, smaller (R.L.M., M.W.V., V.W.V.J.), and Pediatrics (V.W.V.J.), Erasmus Medical Center, Rotter- amygdala volumes and smaller left posterior parahippocampal gyrus dam, the Netherlands; Department of Child and Adolescent Psychiatry (L.M.E.B., 12 R.L.M., H.E.M., F.C.V., H.T., T.W.) and Generation R Study Group (L.M.E.B., R.L.M., volumes compared with patients without CSP were found. Other H.E.M.), Erasmus Medical Center-Sophia, Rotterdam, the Netherlands; Department of Clinical Medicine (F.C.V.), University of Copenhagen, Copenhagen, Denmark; and Harvard School of Public Health (H.T.), Boston, Massachusetts. Please address correspondence to Marjolein Dremmen, MD, Erasmus Medical Cen- ter-Sophia, Department of Radiology, Room Sb-1654, PO Box 2060, 3000 CB Rot- Marjolein H.G. Dremmen and Roos H. Bouhuis contributed equally to this work. terdam, the Netherlands; e-mail: [email protected] This study was financially supported through Netherlands Organization for Health Indicates open access to non-subscribers at www.ajnr.org Research and Development TOP project No. 91211021. The general design of the Generation R Study is made possible by financial support from the Erasmus Medi- Indicates article with supplemental on-line tables. cal Center, Rotterdam; the Erasmus University Rotterdam; the Netherlands Organi- Indicates article with supplemental on-line photos. zation for Health Research and Development; and the Netherlands Organization for Scientific Research, the Ministry of Health, Welfare and Sport. http://dx.doi.org/10.3174/ajnr.A5939

340 Dremmen Feb 2019 www.ajnr.org studies did not find a relation to global or specific brain structure total of 9778 pregnant women were included. Information on the volumes.7,9,13,17 However, because the postnatal closure of the CSP is demographic characteristics included educational levels of the presumed to be dependent on growing brain structures,2,3 we would mother. Multiple measurements were collected during pregnancy expect the brain regions in direct or close contact with the septum (eg, maternal alcohol use and smoking behavior).29 A neuroimaging pellucidum to be smaller in subjects with enlarged CSPs. substudy of children 6–9 years of age was initiated in 2009 and in- Because an enlarged CSP is considered a neurodevelopmental volved 1070 children.28 Exclusion criteria were general contraindica- anomaly, it has been postulated as a potential marker for psychi- tions for MR imaging examination (ie, pacemaker, ferrous metal im- atric disorders that have neurodevelopmental origins. Indeed, an plants), severe motor or sensory disorders (deafness or blindness), enlarged CSP has been evaluated in a broad range of psychiatric neurologic disorders (ie, seizures or tuberous sclerosis), moderate- adult populations, typically by measuring the anterior-to-poste- to-severe head injuries with loss of consciousness, and claustropho- rior CSP length on MR imaging studies, but with mixed results. bia. Informed consent was obtained from the parents before partici- Some studies have shown significantly higher rates of enlarged pation. The study was approved by the Medical Ethics Committee at CSPs in adult patients with schizophrenia,5,9,18 schizophrenia the Erasmus Medical Center-Sophia, Rotterdam.28 spectrum disorder,18 bipolar disorder,11 and disruptive behav- ior disorder19 and opiate-dependent subjects.16 However, MR Imaging Acquisition other studies did not find higher rates of enlarged CSPs in Children were familiarized with the MR imaging scanners using a schizophrenia,2,6,10,12,20 bipolar disorder,21 and other psychi- mock scanning procedure.28 MR images were acquired on a 3T atric disorders, including attention deficit/hyperactivity disor- scanner (Discovery MR750; GE Healthcare, Milwaukee, Wiscon- der,22 borderline personality disorder,13 depression,13 or ob- sin) using an 8-channel head coil for signal reception. Following a sessive-compulsive disorder.14 3-plane localizing and coil-intensity calibration scan, a high-res- The relationship between CSP and cognition is even less clear. olution T1-weighted inversion recovery fast-spoiled gradient-re- Nopoulos et al23 found a negative correlation between cavum called sequence was obtained with the following parameters: length and the intelligence quotient (IQ) in patients with schizo- TR ϭ 10.3 ms, TE ϭ 4.2 ms, TI ϭ 350 ms, NEX ϭ 1, flip angle ϭ phrenia and an enlarged CSP; however, they did not show this 16°, readout bandwidth ϭ 20.8 kHz, matrix ϭ 256 ϫ 256. The relationship between an enlarged CSP and cognition in the con- total scan time for the T1 was 5 minutes 40 seconds.28 The MR trol group. On the other hand, another study has found that a images underwent a 6-parameter affine transformation into a larger proportion of subjects evaluated for mental retardation also study-specific template at 1 ϫ 1 ϫ 1 mm isotropic resolution.30 had an enlarged CSP.24 In addition, an enlarged CSP has been found in children with syndromes that include cognitive impairments such CSP Assessment as fetal alcohol syndrome25 or Apert syndrome.26 These studies may The CSP was assessed in a standardized manner by a single trained suggest that a relationship exists between an enlarged CSP and cog- neuroradiologist, blinded to subject information. Anatomic nitive function that extends to the general population. boundaries for the CSP were as follows: the genu of the corpus The heterogeneity in different cutoffs used to define the pres- callosum defined the anterior boundary, the body of the corpus ence of an enlarged CSP could contribute to some of the mixed callosum defined the superior boundary, the rostrum of the cor- findings. The presence of an enlarged CSP, according to the pus callosum and the defined the inferior Ն6-mm cutoff varied from 0% to 19.5% in patients and from 0% boundary, and the anterior limb and pillars of the defined to 14.7% in controls.2,3,6-16 Because control groups for clinical the posterior boundary. On-line Fig 1 illustrates the anterior and studies do not always reflect the general population, the preva- posterior boundaries of the CSP in the sagittal plane. Similar to lence of an enlarged CSP in the general pediatric population is other studies,5-9,12,13,15,19,20,22,23 the anterior-to-posterior length unknown. Thus, the goal of this study was to determine the prev- of the CSP was measured by counting the number of coronal alence of a CSP in the general pediatric population and to examine 1-mm slices on which a cavum was visible. A cavum was labeled as the relationship between the size of a CSP and brain structure enlarged when its length was Ն6 slices (ie, Ն6 mm).2,3,6-16 Ab- volumes, nonverbal intelligence (as a proxy for cognitive func- sence of a CSP was labeled “no CSP.” A cavum of Ͻ6mmin tion), and emotional or behavioral problems in children. length was categorized as a normal-variant CSP. Figure 1 illus- Given the previously reported link between an enlarged CSP trates the presence of an enlarged CSP. Intrarater reliability was 11,15,19 and behavioral or emotional disorders, we hypothesized based on 50 repeat ratings and was found to be high (test-retest that a relationship between the size of a CSP and behavioral prob- reliability, Cronbach ␣ ϭ 0.96). Scans that were of insufficient lems and cognitive performance would be present in a popula- quality to determine the existence or length of a CSP were ex- tion-based sample of children. cluded from the analyses (12.5%, n ϭ 134). The insufficient scan quality was mainly due to movement artifacts, which are a well- MATERIALS AND METHODS known problem when performing MR imaging studies in unse- Participants dated children at a young age. The current study was embedded in the longitudinal population- based Generation R Study. An overview of the Generation R Study Global and Regional Brain Structures design is published elsewhere.27,28 Briefly, the Generation R Study Volumes of global and regional brain structures were measured is a prospective birth cohort study that started in Rotterdam be- using FreeSurfer image analysis, Suite 5.1 (http://surfer. tween 2002 and 2006. After informed consent was obtained, a nmr.mgh.harvard.edu/). The technical details of these procedures AJNR Am J Neuroradiol 40:340–46 Feb 2019 www.ajnr.org 341 age-specific norms with a mean value of 100 Ϯ 15.36 Handedness was determined using the Edinburgh Handedness Inventory.29

Statistical Analyses Demographic differences between the no/normal-variant CSP (control group) and the enlarged CSP group were analyzed using a ␹2 test for categoric variables (sex, ethnicity, handedness, ma- ternal education, maternal alcohol use during pregnancy, mater- nal smoking behavior during pregnancy) and independent t tests for continuous variables (age). Differences in brain structure vol- umes, nonverbal intelligence, and emotional and behavioral problems between the controls and the enlarged CSP group were tested using ANCOVAs. Covariates were added to the model if they resulted in a Ͼ5% change in the effect estimate, including the child’s age, sex, ethnicity, and maternal education. The analyses of regional brain structure volumes were also adjusted for total problem score and total brain volume. Because the control group did include a small CSP (0–5 mm) and the biologic relevance of a normal-sized CSP is not clear, 2 additional sensitivity analyses were performed for brain structure volumes, nonverbal intelli- FIG 1. MR imaging of the cavum septum pellucidum. Coronal T1- gence, and emotional and behavior problems. First, sensitivity anal- weighted MR image (A) and axial T1-weighted MR image (B) show no yses were performed comparing the 2 extreme categories (0 versus CSP. C, Coronal T1-weighted MR image (C) and axial T1-weighted MR Ն image (D) demonstrate an enlarged CSP. 6 mm), and second, an analysis was performed using a classifica- tion into 3 groups (0 versus 1–5 versus Ն6 mm). For certain analyses of brain structure volumes and emotional or behavioral scores, the have been described elsewhere.31,32 Briefly, this process included length of the CSP was used as a continuous variable. intensity normalization, the removal of nonbrain tissue, auto- All analyses were performed using the Statistical Package for mated Talairach transformation into standard space, and seg- the Social Sciences, Version 21.0 (SPSS; IBM, Armonk, New mentation of the cortical and subcortical white and gray matter York). In addition, a nonparametric analysis with CSP length as a volumetric structures. The volumes of brain regions in close vi- continuous variable was performed using Statistical Analysis Soft- cinity to the CSP were selected for analysis. These regions in- ware, Version 9.3 (SAS Institute, Cary, North Carolina). To cor- cluded the corpus callosum, lateral ventricles, thalamus, hip- rect for multiple testing, we calculated the effective number of pocampus, amygdala, and the caudate nucleus. Volumes of global tests on the basis of the covariance structure among the differ- brain measures were also examined, including total brain, total ent outcomes.37 We found the number of effective tests to be 5; gray matter, and total white matter volumes. thus, the corrected threshold for significance was P Ͻ .01. Data were missing on internalizing symptoms (10.5% missing data), Emotional and Behavioral Problems Emotional and behavioral problems were measured using the externalizing symptoms (10.0%), nonverbal IQ (8.8%), and sum scores of the 99 items of the Child Behavior Checklist the covariates, maternal education (9.5%) and the CBCL sum (CBCL).33 This questionnaire, completed by the parents when the score (9.5%). Missing values were estimated using a multiple im- children were approximately 6 years of age, includes behavioral, putation method with 5 imputations and 10 iterations. Finally, an emotional, and social problems in school-aged children. The additional nonresponse analysis was performed comparing the base- items were calculated for several syndrome scales: emotionally line characteristics of the study participants with the group of chil- reactive, anxious/depressed, somatic complaints, withdrawn, at- dren excluded due to insufficient scan quality. tention problems, and aggressive behavior. Broadband scores of internalizing symptoms and externalizing symptoms were used. RESULTS Sample Characteristics Nonverbal Intelligence Sample characteristics are presented in Table 1. Children in the The Generation R cohort includes children from different ethnic control group and the enlarged CSP group did not differ in age at minorities with differences in Dutch language abilities. Therefore, scanning [t(934) ϭ 0.26, P ϭ .79)], sex [␹2(1) ϭ 0.03, P ϭ .86], nonverbal intelligence was measured using 2 subtests of the ethnicity [␹2(2) ϭ 2.68, P ϭ .26], handedness [␹2(1) ϭ 0.001, P ϭ Snijders-Oomen Nonverbal Intelligence Test 2.5–7-Revised.34 .98], maternal education [␹2(2) ϭ 0.93, P ϭ .63], maternal alco- Due to time constraints, we selected 2 subtests: Mosaics and cat- hol use during pregnancy [␹2(3) ϭ 1.53, P ϭ .68], and maternal egories for testing spatial insight and abstract reasoning abilities, smoking behavior during pregnancy [␹2(2) ϭ 0.23, P ϭ .89]. respectively. Because the correlation between the sum of these 2 subtests and the full SON-R IQ battery is very high (r ϭ 0.86), CSP these subdomain scores can be used as nonverbal IQ scores.35 Raw Figure 2 shows the distribution of the anterior-to-posterior ca- subtest scores were transformed according to population- and vum lengths in the entire sample. The prevalence of a CSP in the 342 Dremmen Feb 2019 www.ajnr.org Table 1: Sample characteristics 0.05, partial ␩2 ϭ 0.004] and larger total Total Group Controls Enlarged CSP white matter/total brain ratio volumes (n = 936) (100%) (n = 893) (95.4%) (n = 43) (4.6%) [F(1,93) ϭ 6.81, P ϭ .009, partial ␩2 ϭ Child characteristics 0.007] compared with the children in Age (mean) (SD) (yr) 7.9 (1.00) 7.9 (1.00) 7.9 (0.91) the control group, adjusting for the co- Sex (%) Girls 45.5 45.6 44.2 variates. After correcting for multiple Boys 54.5 54.4 55.8 testing, only the difference in white mat- Ethnicity (%) ter/total brain ratio volume remained. Dutch 68.8 69.0 65.1 Comparing the children in the controls Other Western 6.8 7.1 2.3 (n ϭ 893) with the those in the enlarged Non-Western 24.4 24.0 32.6 ϭ Handedness (%) CSP group (n 43) or comparison of Right 90.5 90.5 90.7 the 3 groups (0 versus 1–5 versus Ն6 Left 9.4 9.4 9.3 mm) showed similar results. CSP length Maternal characteristics was positively correlated with total brain a Maternal education (%) volume (Spearman partial correlation, Low 10.9 10.8 14.0 ␳ ϭ Ͻ Medium 29.0 29.0 27.9 s 0.19, P .001) and white matter/ High 50.6 51.1 41.9 total brain ratio volume (Spearman par- b ␳ ϭ Ͻ Alcohol during pregnancy (%) tial correlation, s 0.13, P .001). Never 34.3 34.0 39.5 Stopped 13.7 13.8 11.6 Regional Brain Structures Occasionally 35.7 35.7 34.9 Table 3 shows the difference in volumes Frequently 9.2 9.4 4.7 Smoking during pregnancy (%)c of specific structures between the Never 72.2 72.1 74.4 controls and the enlarged CSP group. Stopped 6.3 6.4 4.7 Children in the enlarged CSP group Continued 17.9 17.9 18.6 had larger corpus callosum volumes a Low indicates primary school or lower vocational education; medium, intermediate vocational education; high, higher [F(1,93) ϭ 6.79, P ϭ .009, partial ␩2 ϭ vocational education or university. Missing data on maternal education are 9.5%. b Frequently indicates that the mother drank Ն1 glass per week for at least 2 trimesters. Missing data on alcohol are 7.2%. 0.007], larger thalamus volumes 2 c Smoking Ͼ10 cigarettes per day fluctuated between 3.6% and 5.6% throughout pregnancy, with the highest percent- [F(1,93) ϭ 11.21, P ϭ .001, partial ␩ ϭ age in the first trimester.41 Missing data on smoking are 3.5%. 0.012], and smaller amygdala volumes [F(1,93) ϭ 4.48, P ϭ 0.04, partial ␩2 ϭ 0.005] compared with the children in the control group, after controlling for covariates. After we corrected for multiple testing, only differences in the corpus callosum and thalamus volumes remained. On-line Table 1 presents the results of the comparison of children in the no-CSP group (n ϭ 329) with the those in the enlarged CSP group (Ն6 mm). Additionally, children with an enlarged CSP had smaller lateral ventricle volumes compared with children with complete absence of a cavum [F(1,36) ϭ 8.16, P ϭ .005, partial ␩2 ϭ 0.022]. On-line Table 2 presents the results of the comparison of the 3 groups (0 versus 1–5 versus Ն6 mm). In this analysis, the difference in amygdala volume disappeared, and the 3 groups were found to differ in caudate nucleus volume [F(2,93) ϭ 5.33, P ϭ .005, partial ␩2 ϭ 0.011]. CSP length was FIG 2. Distribution of no CSP, normal-variant CSP, and enlarged CSP positively correlated with the volume of the corpus callosum for the whole sample (n ϭ 936). ␳ ϭ Ͻ (Spearman partial correlation, s 0.12, P .001), thalamus ␳ ϭ ϭ study population was 64.9%, categorized into normal-variant (Spearman partial correlation, s 0.08, P .02), and caudate ␳ ϭϪ ϭ CSP (60.3% of total study population) and enlarged CSP (4.6% of nucleus (Spearman partial correlation, s .10, P .002) and total study population). The mean cavum length in the control was negatively correlated with lateral ventricle volumes (Spear- ␳ ϭϪ Ͻ group was 2.67 mm; lengths ranged from 1 to 5 mm. The mean man partial correlation, s .20, P .001). cavum length in the enlarged CSP group was 12.56 mm; lengths in this group ranged from 6 to 37 mm. A histogram of CSP lengths in Nonverbal Intelligence the enlarged CSP group is shown in On-line Fig 2. Children in the no/normal-sized CSP and the enlarged CSP groups did not differ in nonverbal IQ scores [F(1,93) ϭ 0.05, P ϭ Global Brain Structures 0.82, partial ␩2 ϭ 0.000]. Estimated marginal means (M) for non- Table 2 shows the differences in global structure volumes between verbal IQ were M ϭ 101.63, standard error (SE) ϭ 0.46 versus the controls and the enlarged CSP group. Children in the enlarged M ϭ 101.13, SE ϭ 2.08. The results of comparing the children in CSP group had larger total brain volumes [F(1,93) ϭ 3.93, P ϭ the no-CSP group (n ϭ 329) with the enlarged CSP group (n ϭ AJNR Am J Neuroradiol 40:340–46 Feb 2019 www.ajnr.org 343 Table 2: ANCOVA for global brain structuresa higher CBCL sum scores than the in- Marginal Mean Volumes (cm3) (SE) cluded children [t(970) ϭ 2.53, P ϭ .01]. Controls (n = 893) Enlarged CSP (n = 43) F Sig. Partial ␩2 TBV 1129.09 (4.095) 1167.00 (18.673) 3.93 .05 0.004 DISCUSSION Total GMV/TBV 0.63 (0.001) 0.63 (0.003) 2.14 .14 0.002 To further understand the brain struc- b Total WMV/TBV 0.33 (0.001) 0.34 (0.002) 6.81 .009 0.007 tural correlates associated with an en- Note:—TBV indicates total brain volume; GMV, gray matter volume; WMV, white matter volume; Sig., significance. larged CSP, we evaluated the relation- a Results of the first imputed dataset are reported; ranges of the other datasets can be found in On-line Tables 1 and 2. ship between the CSP and the volume of Covariates are age, sex, ethnicity, maternal education, and CBCL sum score. b Significant after correcting for multiple testing. global and regional brain structures. Children with enlarged CSPs were a Table 3: ANCOVA for regional brain structures (control vs enlarged CSP) found to have larger total white matter/ 3 Marginal Mean Volumes (cm ) (SE) total brain volume ratios and corpus cal- Controls (n = 893) Enlarged CSP (n = 43) F Sig. Partial ␩2 losum and thalamus volumes. Similar to Corpus callosum 2.71 (0.012) 2.86 (0.055) 6.79 .009b 0.007 studies in patients with schizophrenia Lateral ventricles 8.94 (0.172) 7.49 (0.786) 3.22 .07 0.003 spectrum disorder,12 amygdala volumes b Thalamus 14.12 (0.036) 14.69 (0.165) 11.21 .001 0.012 were smaller in subjects with enlarged Hippocampus 8.03 (0.024) 7.94 (0.111) 0.67 .41 0.001 Amygdala 3.17 (0.014) 3.04 (0.062) 4.48 .04 0.005 CSPs. However, this result did not re- Caudate nucleus 8.44 (0.032) 8.30 (0.148) 0.91 .34 0.001 main significant after correction for Note:—Sig. indicates significance. multiple testing. In addition, children a Results of the first imputed dataset are reported; ranges of the other datasets can be found in On-line Tables 1 and 2. with enlarged CSPs had smaller lateral Covariates are age, sex, ethnicity, total brain volume, maternal education, and CBCL sum score. ventricle volumes compared with chil- b Significant after correcting for multiple testing. dren with a total absence of a cavum. CSP length was positively correlated 43) were also not significant. Comparing 3 groups (0 versus 1–5 with total brain volume; white matter/total brain ratio; and cor- versus Ն6 mm) did show a difference in nonverbal IQ scores pus callosum, thalamus, and caudate nucleus volumes. CSP [F(1,93) ϭ 3.37, P ϭ 0.04, partial ␩2 ϭ 0.007]. Estimated mar- length was negatively correlated with lateral ventricle volumes. ginal means for the 3 groups were M ϭ 100.08, SE ϭ 0.75 versus It has been postulated that the postnatal closure of the CSP is a M ϭ 102.53, SE ϭ 0.57 versus M ϭ 101.14, SE ϭ 2.07; however, result of growth of surrounding brain structures, exerting pres- this finding did not survive correction for multiple testing. A non- sure on the 2 leaflets of the septum pellucidum and thereby facil- parametric test with CSP length as a continuous variable did not itating fusion.2,3 Brain development in this early postnatal period show a correlation between CSP length and nonverbal IQ scores. has not been extensively studied, but a recent study on infant brain development in the first 3 months after birth38 presents Emotional or Behavioral Problems growth rates for several ROIs of the current study. The lateral Children in the control group and the enlarged CSP group did not ventricles were found to grow at the highest rate, increasing differ in emotional [F(1,93) ϭ 2.35, P ϭ 0.13, partial ␩2 ϭ 0.003] around 78% in the first 90 days after birth. As hypothesized, the or behavioral [F(1,93) ϭ 0.35, P ϭ 0.55, partial ␩2 ϭ 0.000] prob- rapid growth of the lateral ventricles in these first months after lem scores. Estimated marginal means for emotional scores were birth could induce closure of the cavum by exerting lateral pres- M ϭ 2.48, SE ϭ 0.04 versus M ϭ 2.18, SE ϭ 0.19. Estimated sure on both sides of the septal leaves. We speculated that smaller marginal means for behavioral scores were M ϭ 2.80, SE ϭ 0.04 growth rates in this period could cause failure of CSP closure. If versus M ϭ 2.93, SE ϭ 0.20. Mean scores are reported as square- these early growth rates are decisive for structure volumes at a root-transformed scores. Comparing the children in the no-CSP later age, our results support this hypothesis because the lateral group (n ϭ 329) with those in the enlarged CSP group (n ϭ 43) or ventricles were smaller in children with the presence of a CSP comparing 3 groups (0 mm 1–5 versus Ն6 mm) did not change compared with those with total absence of a cavum. Previous the results. A Spearman correlation coefficient test with CSP studies have reported an association between an enlarged CSP and length as a continuous variable also did not show a correlation between CSP length and emotional or behavioral scores. syndromes including cognitive impairment such as fetal alcohol syndrome and Apert syndrome.25,26 Participant versus Excluded Analysis Mechanical factors that potentially play a role in the closure of Of all 1070 scans, 134 scans (12.5%) were of insufficient quality the CSP in the general pediatric population could also extend to and were excluded from the analyses. Children whose scans were these clinical conditions. In these syndromes, there is an obvious excluded did not differ from the current study sample (n ϭ 936) change in brain volume and/or head size, which could cause ma- in age at the time of scanning [t(1068) ϭϪ0.81, P ϭ .42], sex jor changes in the exerting pressure on the septal leaves, leading to [␹2(1) ϭ 3.18, P ϭ .07], ethnicity [␹2(2) ϭ 3.85, P ϭ .15], hand- a failure of CSP closure. The enlarged CSP would rather be a edness [␹2(1) ϭ 0.43, P ϭ .51], nonverbal IQ [t(980) ϭϪ0.59, consequence of the anatomic and structural abnormalities in P ϭ .56], maternal education [␹2(2) ϭ 3.83, P ϭ .15], maternal these syndromes than a marker of cognitive function itself. In alcohol use during pregnancy [␹2(3) ϭ 1.12, P ϭ .77], and ma- addition, the larger total amount of white matter (larger total ternal smoking behavior during pregnancy [␹2(2) ϭ 5.64, white matter/total brain volumes ratios) in the enlarged CSP P ϭ .06]. However, children excluded from the analyses had group could also contribute to a change in mechanical factors that 344 Dremmen Feb 2019 www.ajnr.org cause the lateral ventricles to grow less rapidly in the early post- There are some limitations in the study protocol. Only 1 rater natal period. This, in turn, could modify the exerting lateral pres- evaluated the MR imaging studies and assessed the presence and sure on both sides of the septal leaves and alter the possible closure size of the CSP. However, the rater was specifically trained to mechanisms of the septal leaves. evaluate the CSP and performed all the measurements in a stan- In addition, other regional brain structures were also found to dardized manner. Furthermore, 1 exclusion criterion to undergo be larger in children with a persistent CSP. For example, the cor- MR imaging examination in the Generation R Study design is a pus callosum is a structure that directly surrounds the CSP but history of moderate-to-severe head injuries with loss of con- was found to be larger in children with an enlarged CSP. We sciousness. In previous literature, a possible relationship between speculated that the exertion of lateral pressure, like the pressure of an enlarged CSP and traumatic brain injury has been reported in growing lateral ventricles, is more effective in facilitating the post- adults as well as in children.39,40 Therefore, exclusion of children natal fusing process of the septum leaves if the corpus callosum is with previous moderate-to-severe brain injury potentially re- smaller. Thus, under the rubric of a mechanical closure hypothe- sulted in selection bias. In addition, missing values on internaliz- sis of the CSP, larger brain structures would require greater dis- ing and externalizing symptoms, nonverbal IQ, and certain cova- tributed pressure along the structure to close the septum. How- riates are a potential limitation of the study design, but we applied ever, because we do not have MR imaging data available during multiple imputation methods to reduce the potential bias intro- the early postnatal period, it is not possible to draw conclusions duced as a result of missing data to a minimum. concerning the timing and longitudinal relationship with neigh- The strengths of the current study are the standardized MR boring structures related to the postnatal closure of the CSP. It imaging measurements obtained in a large population-based is likely that multiple-yet-unknown factors play a role in this sample of 1070 children of different ethnicities and backgrounds. process. The study contains an extensive data collection with a broad range Furthermore, it would be interesting to search for genetic fac- of physiologic and environmental measures. Therefore, it in- tors that play a role in the closure of the CSP. Because the Gener- cludes valuable data representative of the pediatric population in ation R Study is currently starting to prospectively scan the par- general. ents of a subgroup of children scanned in this study, assessment of the presence and size of a CSP in the parents could be of great CONCLUSIONS value to demonstrate the influence of genetics. This may well be a The CSP is a structural brain variation without cognitive or be- focus of further studies. havioral implications in this population-based sample of school- The prevalence of an enlarged CSP in our population-based aged children. However, we did find a relationship between CSP sample of school-aged children is 4.6%. This falls within the range and global and regional volumetric brain measures, but the clin- described in prior literature in adult populations.2,3,5-16 While ical relevance of this relationship is as yet uncertain. earlier research tended to focus on psychiatric populations, the prevalence in healthy control groups in these studies varied sub- Disclosures: Marjolein H.G. Dremmen—RELATED: Grant: the Netherlands Organiza- tion for Health Research and Development*; UNRELATED: Employment: Pediatric stantially from 0% to 14.7%. Because many of these studies had Radiologist at Erasmus Medical Center, Rotterdam. Roos H. Bouhuis—RELATED: small sample sizes and control groups are not always reflective of Grant: the Netherlands Organization for Health Research and Development.* Hanan El Marroun—RELATED: Grant: European Union, Comments: The Horizon 2020 re- recruitment from the general population, we believe that our search and innovation program (grant agreement No.633595 DynaHEALTH and study provides more accurate rates of enlarged CSPs in the general No.733206 LifeCycle) of the European Union.* Frank C. Verhulst—UNRELATED: Roy- pediatric population. However, we are unaware of studies assess- alties: distributor of the Dutch translations of the Achenbach System of Empirically Based Assessment, from which I receive remuneration. Tonya White—RELATED: ing changes with time in the size or presence of the CSP; thus, Grant: the Netherlands Organization for Health Research and Development.* longitudinal studies are necessary to address this question. *Money paid to the institution. An enlarged CSP is considered a potential marker for neuro- developmental abnormalities. However, in contrast to earlier REFERENCES 11,15,19 studies, we did not find a relationship between an enlarged 1. Sarwar M. The septum pellucidum: normal and abnormal. AJNR CSP and nonverbal intelligence. We also did not find a relation- Am J Neuroradiol 1989;10:989–1005 Medline ship between emotional or behavioral problems and an enlarged 2. Dickey CC, McCarley RW, Xu ML, et al. MRI abnormalities of the hippocampus and cavum septi pellucidi in females with schizotypal CSP. While we found no evidence for behavioral correlates of an personality disorder. Schizophr Res 2007;89:49–58 CrossRef Medline enlarged CSP in this population-based sample of school-aged 3. Choi JS, Kang DH, Park JY, et al. Cavum septum pellucidum in children, some neuropsychiatric disorders that show a relation- subjects at ultra-high risk for psychosis: compared with first-de- ship with an enlarged CSP have a typical age of onset in late ado- gree relatives of patients with schizophrenia and healthy volun- lescence and early adulthood. For example, schizophrenia spec- teers. Prog Neuropsychopharmacol Biol Psychiatry 2008;32:1326–30 CrossRef Medline trum disorders and bipolar disorders generally present with the 4. Shunk H. Congenital dilatations of the septi pellucidum. Radiology 7,21 first symptoms between 17 and 25 years of age. Thus, it is 1963;81:610–18 CrossRef Medline possible that the relationship between an enlarged CSP and psy- 5. Nopoulos PC, Giedd JN, Andreasen NC, et al. Frequency and sever- chopathology will emerge later, parallel to the emergence of psy- ity of enlarged cavum septi pellucidi in childhood-onset schizo- chiatric symptoms. The small, nonsignificant, volumetric differ- phrenia. Am J Psychiatry 1998;155:1074–79 CrossRef Medline 6. Hagino H, Suzuki M, Kurokawa K, et al. Magnetic resonance imag- ences of certain brain structures found in this study could ing study of the cavum septi pellucidi in patients with schizophre- theoretically indicate some structural changes in brain regions nia. Am J Psychiatry 2001;158:1717–19 CrossRef Medline that might form the basis for these diseases. 7. Rajarethinam R, Miedler J, DeQuardo J, et al. Prevalence of cavum AJNR Am J Neuroradiol 40:340–46 Feb 2019 www.ajnr.org 345 septum pellucidum in schizophrenia studied with MRI. Schizophr cavum vergae in normal and developmentally delayed populations. Res 2001;48:201–05 CrossRef Medline J Child Neurol 1998;13:120–21 CrossRef Medline 8. Crippa JA, Uchida R, Busatto GF, et al. The size and prevalence of the 25. Swayze VW 2nd, Johnson VP, Hansen JW, et al. Magnetic resonance cavum septum pellucidum are normal in subjects with panic disor- imaging of brain anomalies in fetal alcohol syndrome. Pediatrics der. Braz J Med Biol Res 2004;37:371–74 CrossRef Medline 1997;99:232–40 CrossRef Medline 9. de Souza Crippa JA, Zuardi AW, Busatto GF, et al. Cavum septum 26. Renier D, Arnaud E, Cinalli G, et al. Prognosis for mental function in pellucidum and adhesio interthalamica in schizophrenia: an MRI Apert’s syndrome. J Neurosurg 1996;85:66–72 Medline study. Eur Psychiatry 2006;21:291–99 CrossRef Medline 27. Jaddoe VW, Mackenbach JP, Moll HA, et al. The Generation R Study: 10. Flashman LA, Roth RM, Pixley HS, et al. Cavum septum pellucidum design and cohort profile. Eur J Epidemiol 2006;21:475–84 CrossRef in schizophrenia: clinical and neuropsychological correlates. Psy- Medline chiatry Res 2007;154:147–55 CrossRef Medline 28. White T, El Marroun H, Nijs I, et al. Pediatric population-based 11. Kim MJ, Lyoo IK, Dager SR, et al. The occurrence of cavum septi neuroimaging and the Generation R Study: the intersection of de- pellucidi enlargement is increased in bipolar disorder patients. Bi- velopmental neuroscience and epidemiology. Eur J Epidemiol. 2013; polar Disord 2007;9:274–80 CrossRef Medline 28:99–111 CrossRef Medline 12. Takahashi T, Suzuki M, Hagino H, et al. Prevalence of large cavum 29. White T, Muetzel RL, El Marroun H, et al. Paediatric population septi pellucidi and its relation to the medial temporal lobe struc- neuroimaging and the Generation R Study: the second wave. Eur J tures in schizophrenia spectrum. Prog Neuropsychopharmacol Biol Epidemiol 2018;33:99–125 CrossRef Medline Psychiatry 2007;31:1235–41 CrossRef Medline 30. Muetzel RL, Blanken LME, Thijssen S, et al. Variability in ICA-de- 13. Takahashi T, Yu¨cel M, Lorenzetti V, et al. Midline brain structures in rived resting-state networks in 6-to-10 year-old children. Hum patients with current and remitted major depression. Prog Neuro- Brain Mapp 2016;37:4286–4300 CrossRef Medline psychopharmacol Biol Psychiatry 2009;33:1058–63 CrossRef Medline 31. White T, Jansen PR, Muetzel RL, et al. Automated quality assess- 14. Chon MW, Choi JS, Kang DH, et al. MRI study of the cavum septum ment of structural magnetic resonance images in children: compar- ison with visual inspection and surface-based reconstruction. Hum pellucidum in obsessive-compulsive disorder. Eur Arch Psychiatry Brain Mapp 2018;39:1218–31 CrossRef Medline Clin Neurosci 2010;260:337–43 CrossRef Medline 32. Fischl B, Salat DH, Busa E, et al. Whole brain segmentation: auto- 15. Raine A, Lee L, Yang Y, et al. Neurodevelopmental marker for limbic mated labeling of neuroanatomical structures in the . maldevelopment in antisocial personality disorder and psychopa- Neuron 2002;33:341–55 CrossRef Medline thy. Br J Psychiatry 2010;197:186–92 CrossRef Medline 33. Achenbach TM, Rescorla LA. Manual for the ASEBA Preschool Forms 16. Hwang J, Kim JE, Kaufman MJ, et al. Enlarged cavum septum pellu- and Profiles. Burlington, VT: University of Vermont, Research Center cidum as a neurodevelopmental marker in adolescent-onset opiate for Children, Youth, & Families; 2000 dependence. PLoS One 2013;8:e78590 CrossRef Medline 34. Basten M, van der Ende J, Tiemeier H, et al. Nonverbal intelligence in 17. Galarza M, Merlo AB, Ingratta A, et al. Cavum septum pellucidum young children with dysregulation: the Generation R Study. Eur and its increased prevalence in schizophrenia: a neuroembryologi- Child Adolesc Psychiatry 2014;23:1061–70 CrossRef Medline cal classification. J Neuropsychiatry Clin Neurosci 2004;16:41–46 35. Moore C, O’Keefe S, Lawhorn D, et al, et al. Concurrent validity of CrossRef Medline the Snijders-Oomen Nonverbal Intelligence Test 2 1/2–7–Revised 18. Kasai K, McCarley RW, Salisbury DF, et al. Cavum septi pellucidi in with the Wechsler Preschool and Primary Scale of Intelligence– first-episode schizophrenia and first-episode affective psychosis: Revised. Psychological Reports 1998;82:619–25 CrossRef an MRI study. Schizophr Res 2004;71:65–76 CrossRef Medline 36. Snijders JH, Snijders-Oomen AW, Tellegen PJ, et al. SON-R 21/2–7: 19. White SF, Brislin S, Sinclair S, et al. The relationship between large Snijders-Oomen Niet-Verbale Intelligentietest—Handleiding en Ve- cavum septum pellucidum and antisocial behavior, callous-un- rantwoording. Amsterdam: Boom Testuitgevers; 2005 emotional traits and psychopathy in adolescents. J Child Psychol 37. Galwey NW. A new measure of the effective number of tests, a prac- Psychiatry 2013;54:575–81 CrossRef Medline tical tool for comparing families of non-independent significance 20. Rajarethinam R, Sohi J, Arfken C, et al. No difference in the preva- tests. Genet Epidemiol 2009;33:559–68 CrossRef Medline lence of cavum septum pellucidum (CSP) between first-episode 38. Holland D, Chang L, Ernst TM, et al. Structural growth trajectories schizophrenia patients, offspring of schizophrenia patients and and rates of change in the first 3 months of infant brain develop- healthy controls. Schizophr Res 2008;103:22–25 CrossRef Medline ment. JAMA Neurol 2014;71:1266–74 CrossRef Medline 21. Takahashi T, Malhi GS, Wood SJ, et al. Midline brain abnormalities 39. Lee JK, Wu J, Banks S, et al. Prevalence of traumatic findings on in established bipolar affective disorder. J Affect Disord 2010;122: routine MRI in a large cohort of professional fighters. AJNR Am J 301–05 CrossRef Medline Neuroradiol 2017;38:1303–10 CrossRef Medline 22. Nopoulos P, Berg S, Castellenos FX, et al. Developmental brain 40. Silk T, Beare R, Crossley L, et al. Cavum septum pellucidum in anomalies in children with attention-deficit hyperactivity disorder. pediatric traumatic brain injury. Psychiatry Res 2013;213:186–92 J Child Neurol 2000;15:102–08 CrossRef Medline CrossRef Medline 23. Nopoulos P, Krie A, Andreasen NC. Enlarged cavum septi pellucidi 41. El Marroun H, White TJ, van der Knaap NJ, et al. Prenatal exposure in patients with schizophrenia: clinical and cognitive correlates. to selective serotonin reuptake inhibitors and social responsiveness J Neuropsychiatry Clin Neurosci 2000;12:344–49 CrossRef Medline symptoms of autism: population-based study of young children. 24. Bodensteiner JB, Schaefer GB, Craft JM. Cavum septi pellucidi and Br J Psychiatry 2014;205:95–102 CrossRef Medline

346 Dremmen Feb 2019 www.ajnr.org